Corpus Content

This corpus will represent playlists with generally similar songs, but that are or are not classified as ‘singing in the shower’ playlists. By comparing these groups, I will try to see if there is a measurable difference in certain attributes measured by Spotify, to define what it means to be a ‘singing in the shower’ playlist. I chose the most-followed ‘singing in the shower’ playlists on Spotify, so I believe it should be a good representation of music that people do enjoy while showering. For more information on the playlists used, take a look at the Appendix storyboard.

In later storyboards, I will also be examining one outlier from each of these playlists: ‘7 rings’ by Ariana Grande (singing in the shower) and ‘Perfect’ by Ed Sheeran (pop), in order to visualize some of the differences between the playlists, since both playlists are quite similar.

Examining the Energy and Valence for Both Sets of Playlists

     Looking at this first scatterplot, we see the greatest concentration of music, for both playlists, in the quadrant with high energy and high valence. This seems valid, given both playlists seem to have a large mix of fast-paced, loud, and cheerful music. As such, we can see our first real similarity in both playlists.

     More interestingly, we see that the ‘singing in the shower’ playlists have a wider array of music. Particularly, it seems to have a decent proportion of songs with 0-0.5 valence, as compared to the ‘pop’ playlist. One theory for this is that top charts ‘pop’ music is generally “catchy” because it has a positive and fast beat, but songs that people generally like to sing along to can be happy or more angry, that usually have fast and loud beats. Neither, however, seem to have much music with low energy and low valence, but music in that category tends to be perceived as sadder, which makes it less likely to be considered a ‘pop’ song, and also most likely makes it not as fun to sing in the shower (though this is mere speculation on my behalf).

Comparing the Danceability and Loudness for Both Sets of Playlists

• Most of the music from both playlists is concentrated in the same area.
• Danceability: ~0.6.
• Loudness: ~-5.
• ‘Singing in the Shower’ playlists seem to have a greater range in terms of both danceability and loudness.
• Possible explanation: ‘pop’ songs that are generally the most enjoyed are ones that are loud, and also that are good to dance to (both at clubs/parties, and at home), music that is enjoyed while showering may not need to be good to dance along to (since there is generally not space to dance in the shower), but mostly on the louder side, so that people can still “belt out tunes” while showering.
• Distribution of data seems valid.
• Further research is necessary to paint a clearer picture of defining ‘singing in the shower’ characteristics.

Comparing Each Playlist’s Top Song Chroma Features

Looking at the most popular song from each set of playlists:

• Shower - 7 rings by Ariana Grande (Manhattan)
• Pop - Perfect by Ed Sheeran (Euclidean)

In a preliminary analysis of the plots, we can see that ‘Perfect’ has more dense chroma features. It also has noticeable use of pitch classes Bb, F, and Eb, unlike ‘7 rings’. On the other hand, ‘7 rings’ uses pitch classes A and F#, unlike ‘Perfect’. It’s interesting to see how different the most popular song from each playlist is just based on chroma values.

Comparing Each Playlist’s Top Song Chroma and Timbre Self-Similarity Matrices

The differences noted in the previous storyboard between the top song from each playlist are even more noticeable when comparing the chroma and timbre self-similarities matrices of both songs. The most noticeable difference is that ‘Perfect’ has more points of novelty / changes throughout the song, whereas ‘7 rings’ has few unexpected parts of the song (where the biggest change is at the start of the song). Rather, ‘7 rings’ is far more homogenous. That said, both songs have few texture changes, and have mostly consistent song structures, if you exclude the beginning of both songs.

These songs are outliers in terms of popularity, but also in terms of other track-level features (energy, valence, etc.). That said, I chose these songs for my comparisons because I thought they would be interesting to highlight some of the differences between the two sets of playlists, since both are pretty similar when looked at through other measures.

Comparing Average Timbre Coefficients for Both Sets of Playlists

Timbre coefficients 3, 5, and 7 seem like the best markers for differences in both sets of playlists. For more context, the third looks at the flatness of the songs, the fifth looks at the brightness of the song at different points of time, and the seventh looks at a combination of brightness and songs with a beginning attack.

That said, the differences (especially when looking at coefficients 3 and 5) don’t seem to be that significant. Also, the rest of the coefficients seem to indicate close similarities in timbre values across both sets of playlists, suggesting how similar the ‘singing in the shower’ playlists are to the pop ones.

Appendix

Singing in the Shower Playlists:

Pop Playlists: